The Journal of Infectious Diseases SUPPLEMENT ARTICLE

Enteric Fever Cases in the Two Largest Pediatric Hospitals of Bangladesh: 2013–2014 Shampa Saha,1 Mohammad J. Uddin,1 Maksuda Islam,1 Rajib C. Das,1,a Denise Garrett,2 and Samir Kumar Saha1,3, 1Child Health Research Foundation, Department of , Dhaka Shishu (Children) Hospital, Bangladesh; 2Sabin Vaccine Institute, Washington, DC; 3Bangladesh Institute of Child Health, Dhaka Shishu (Children) Hospital, Bangladesh

Background. Enteric fever predominantly affects children in low- and middle-income countries. This study examines the bur- den of enteric fever at the 2 pediatric hospitals in Dhaka, Bangladesh and assesses their capacity for inclusion in a prospective cohort study to support enteric fever prevention and control. Methods. A descriptive study of enteric fever was conducted among children admitted in 2013–2014 to inpatient departments of Dhaka Shishu and Shishu Shashthya Foundation Hospitals, sentinel hospitals of the World Health Organization-supported Invasive Bacterial Vaccine Preventable Disease surveillance platform. Results. Of 15 917 children with blood specimens received by laboratories, 2.8% (443 of 15 917) were culture positive for signif- icant bacterial growth. Sixty-three percent (279 of 443) of these isolates were confirmed as the cases of enteric fever (241Salmonella Typhi and 38 Salmonella Paratyphi A). In addition, 1591 children had suspected enteric fever. Overall, 3.6% (1870 of 51 923) were laboratory confirmed or suspected enteric fever cases (55% male, median age 2 years, 86% from Dhaka district, median hospital stay 5 days). Conclusions. The burden of enteric fever among inpatients at 2 pediatric hospitals in Dhaka, Bangladesh is substantial. Therefore, inclusion of these hospitals in a prospective cohort study will be useful for the generation of credible disease burden estimates of enteric fever in Bangladesh. Keywords. Bangladesh; enteric fever; hospital; Salmonella; surveillance.

Enteric fever is an acute, systemic infectious disease caused The establishment of a comprehensive surveillance system is by the Salmonella enterica serovars Typhi and important for generating credible data on enteric fever disease Paratyphi A, B, or C, which often manifests with high-grade burden in endemic regions [10]. In Bangladesh, there was no fever, coated tongue, nausea or vomiting, diarrhea, abdom- comprehensive surveillance that systematically collected infor- inal pain, and cough [1]. It predominantly affects children mation on enteric fever for measuring burden of the disease. and young adults because they either lack natural immunity To date, the burden of enteric fever in Bangladesh has been or experience high levels of exposure to fecal pathogens [2, 3]. estimated on the basis of studies conducted in an urban slum STANDARD Each year, approximately 16 million cases of illness and over [11, 12]. Some hospital-based studies that estimated burden 153 000 deaths are attributed to enteric fever, although esti- of enteric fever included hospitalized cases only [13, 14]. The mates vary and are uncertain due to the limited number of pop- lack of credible estimates for enteric fever incidence, compli- ulation-based incidence studies [4]. In addition, enteric fever cations, and mortality rates has hampered the establishment continues to be an important global health problem especially of evidence-based policy decisions to prevent and control this in low- and middle-income countries of South Asia, including disease [15]. Bangladesh [2, 4–8]. These low-resource countries experience a To fill these knowledge gaps, the Surveillance for Enteric high burden of enteric fever because they have limited access to fever in Asia Project (SEAP), a phased, comprehensive, mul- safe drinking water and to adequate sanitation and hygiene [2, 9]. ticountry, and multisite study, was initiated. The SEAP’s over- all objectives are to characterize the burden of enteric fever in selected Asian settings, including clinical manifestations, sever- aPresent Affiliation: Evaluation Officer, Bangladesh Handloom Board (Ministry of Textiles and ity of illness, long-term sequelae of illness, antimicrobial resis- Jute), 7–9 Kawran Bazar. Correspondence: S. K. Saha, PhD, Executive Director, Child Health Research Foundation, and tance patterns of Salmonella Typhi and Paratyphi, and cost of Professor and Head, Department of Microbiology, Bangladesh, Institute of Child Health, Dhaka illness. The SEAP study will generate baseline data to inform Shishu Hospital, Sher-E-Bangla Nagar, Dhaka-1207, Bangladesh ([email protected]). appropriate interventions for enteric fever prevention and con- The Journal of Infectious Diseases® 2018;218(S4):S195–200 © The Author(s) 2018. Published by Oxford University Press for the Infectious Diseases Society trol and facilitate the assessment of the impact of interventions of America. This is an Open Access article distributed under the terms of the Creative Commons while also characterizing risk factors for the development of Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. severe illness among enteric fever cases. Phase I of the SEAP DOI: 10.1093/infdis/jiy521 was aimed at collecting data on enteric fever cases in 4 countries

Enteric Fever Among Children of Urban Dhaka • JID 2018:218 (Suppl 4) • S195 (Bangladesh, India, Pakistan, and Nepal) to assess health facil- Detection of Salmonella Typhi and Salmonella Paratyphi in Blood ities with the potential to participate in and guide the design of Samples Phase II of the project, a prospective surveillance study to be All blood cultures were performed using standard methods conducted at selected hospitals and laboratory networks in 3 [16]. In briefly, 2–3 mL blood was obtained under aseptic con- endemic countries, namely, Bangladesh, Nepal, and Pakistan. dition and inoculated into Trypticase soy broth supplemented In this study, we provide descriptive data from the SEAP Phase with sodium polyethanol sulphonate (0.25%) and isovitalex I to assess the burden of enteric fever at 2 pediatric hospitals in (1%). After incubation, blood culture bottles were subcultured Dhaka, Bangladesh as well as evaluate their capacity to partici- on second, third, and fifth days of incubation. Standard bio- pate in a prospective cohort study. chemical tests and agglutination with Salmonella species and serovar specific antisera (Ramel; Thermo Fisher Scientific) METHODS were used to confirmSalmonella Typhi/Paratyphi isolates. To Study Design, Sites, and Procedures ensure the quality of the laboratories, we have both internal and Data for the SEAP Phase I were gathered retrospectively from external quality control (QC) systems in place. Internal QC of an ongoing surveillance project during 2013–2014 among inpa- the laboratories is performed at quarterly basis. External QC is tient departments (IPDs) of 2 hospitals in Dhaka, Bangladesh: performed once a year by United Kingdom National External (1) Dhaka Shishu (Children) Hospital (DSH) and (2) Shishu Quality Assessment Scheme as part of QC of IB-VPD surveil- Sashthya (Child Health) Foundation Hospital (SSFH). The DSH lance. In addition, we send specimens to the Center for Disease is the largest pediatric hospital in Bangladesh that provides pri- Control and Prevention to determine whether the results were mary, secondary, and tertiary care to patients less than 18 years concordant. Blood culture volume is regularly monitored in the of age; 37% of the 640 beds in DSH are reserved for patients laboratories. who are unable to pay. With 200 beds, SSFH is the second larg- Data Analysis est pediatric (patients <18 years) hospital in the country and All analyses were performed using the STATA 13.1 (StataCorp provides primary care only; 5% of admitted patients are pro- LP, College Station, TX). Summary statistics included vided care free of charge. In both hospitals, blood specimens are median ± interquartile range (IQR) for continuous variables collected at the treating physician’s discretion and processed in and frequencies with percentages for categorical variables. the in-house microbiology laboratories. Comparative statistics included independent samples t test, These 2 hospitals are sentinel sites of the World Health Wilcoxon’s rank-sum test, χ2 test, and Fisher’s exact test, as Organization (WHO)-coordinated Invasive Bacterial Vaccine appropriate. Two-sided statistical tests were conducted at an Preventable Disease (IB-VPD) surveillance. In our ongoing alpha level of 0.05. WHO-supported IB-VPD surveillance at DSH and SSFH, we collect detailed demographic, clinical, and laboratory informa- Ethical Consideration tion electronically on hospitalized suspected and laboratory-con- Approval for analysis of secondary data for SEAP Phase I was firmed , sepsis, , and enteric fever cases who obtained from the Ethical Review Committee of Bangladesh are younger than 5 years. The inclusion criteria for enteric fever is Institute of Child Health, Dhaka, with a waiver of informed ≥38.9°C for ≥3 days [6]. Minimal demographic and clinical infor- consent. mation (for example, age, sex, diagnoses on admission and dis- charge, hospital duration, etc) of the hospitalized children who are RESULTS not enrolled in IB-VPD surveillance are recorded in the hospital Enteric Fever Cases Identified at Each Site database. For the present study, we extracted demographic (age, Between January 2013 and December 2014, a total of 51 923 sex, address), clinical (diagnoses, symptoms, duration of hospital- children were admitted in the IPDs of the 2 study hospitals ization, final outcome), and laboratory data of enteric fever cases (44 111 at DSH and 7812 at SSFH) (Table 1 and Figure 1). The from the IB-VPD surveillance and hospital databases. age of the hospitalized children ranged from 1 day to 18 years Case Definition with a median of 7 months (IQR, <1 month–24 months). The A suspected case of enteric fever was defined as a patient with majority of children (88.2%, 45 790 of 51 923) admitted to these a final clinical diagnosis of enteric fever without laboratory hospitals were less than 5 years of age, with 73.9% (38 364 of confirmation of Salmonella Typhi/Paratyphi through 51 923) younger than 2 years. blood culture. This final diagnosis was based on physician’s Of 51 923 children admitted to these 2 hospitals, 15 917 had assessment at the end of patient follow-up (ie, discharge/death/ blood specimens collected and received by laboratories for referral/left against medical advice). A laboratory-confirmed culture, and 2.8% (443 of 15 917) of these specimens were cul- case of enteric fever was defined as a patient whose blood cul- ture positive for significant microbiological growth. The total ture was positive for Salmonella Typhi/Paratyphi. number of patients eligible for blood draw was not available.

S196 • JID 2018:218 (Suppl 4) • Saha et al Table 1. Total Admissions, Blood Culture Performed, and Enteric Fever Cases by Study Hospital, Dhaka, Bangladesh, 2013–2014 (n = 51 923)

Shishu Shasthya Dhaka Shishu Hospital Foundation Hospital Total

Parameters n % n % n %

Total admissions 44 111 7812 51 923 Blood culture performed 11 888 27.0 4029 51.6 15 917 30.7 Blood culture positive, any organism 289 2.4 154 3.8 443 2.8 Blood culture positive, enteric fever 151 1. 3 128 3.2 279 1. 8 Salmonella Typhi 135 89.4 106 82.8 241 86.4 Salmonella Paratyphi A 16 10.6 22 1 7. 2 38 13.6

Furthermore, 1.8% of all blood cultures (279 of 15 917) or 63% hospitalized children who either had no blood culture done or of culture-positive cases (279 of 443) were laboratory-con- whose blood culture was negative for microbiological growth firmed enteric fever cases; 86% (241 of 279) of them were lab- (n = 51 480), 1591 had a final clinical diagnosis of enteric fever. oratory-confirmed Salmonella Typhi, and 14% (38 of 279) were In total, 3.6% (1870 of 51 923) of all admissions at the study Salmonella Paratyphi A. Approximately 83% (128 of 154) of all hospitals were either laboratory-confirmed or suspected enteric isolates in SSFH were positive for Salmonella Typhi of Paratyphi fever cases; the proportion of laboratory-confirmed and sus- A, whereas 52% (151 of 289) of all isolates in DSH were positive pected enteric fever cases was significantly higher (P < .001) in for Salmonella Typhi/Paratyphi A (Table 1 and Figure 1). Among SSFH (9.1%, 711 of 7812) than in DSH (2.6%, 1159 of 44 111).

Total admissions N = 51 923

Blood culture not performed N = 36 006 Suspected case Blood culture performed (Enteric fever) N = 15 917 N = 1591

Blood culture negative for significant organism N = 15 474

Blood culture positive for significant organism N = 443

Blood culture positive for other organisms N = 164

Blood culture positive for Salmonella Typhi/Paratyphi Confirmed case (Enteric fever) N = 279

Salmonella Typhi Salmonella Paratyphi N = 241 N = 38

Figure 1. Flowchart for the identification of enteric fever cases at the study hospitals, Dhaka, Bangladesh, 2013–2014.

Enteric Fever Among Children of Urban Dhaka • JID 2018:218 (Suppl 4) • S197 350 100% 90% 300 80% 250 70%

r 200 60% 50% 150 Numbe 40% Percentage 100 30% 20% 50 10% 0 0% <1 123456789101112131415 Age in year

Suspected casesLab-confirmed cases

Cumulative % (Suspected) Cumulative % (Lab-confirmed)

Figure 2. Age distribution of suspected and laboratory-confirmed enteric fever cases who were admitted at the study hospitals, Dhaka, Bangladesh, 2013–2014 (n = 1870).

Characteristics of Suspected and Laboratory-Confirmed Enteric identified among infants younger than 7 months. The median Fever Cases age of suspected and laboratory-confirmed enteric fever cases Among all suspected and laboratory-confirmed enteric fever at both hospitals was 3 (IQR, 1–5) years. Compared with SSFH cases, 55% (1032 of 1870) were male. The majority (86%, 1606 (median, 3; IQR, 2–6 years), admitted suspected and laborato- of 1870) of cases lived within the Dhaka district. Detailed infor- ry-confirmed enteric fever cases were younger at DSH (median, mation on residential address was available for 61% (1133 of 2; IQR, 1–4 years; P < .001). Laboratory-confirmed cases were 1870) of cases. Of them, 83% (946 of 1133) lived within Dhaka; significantly more likely to live within the Dhaka area and had 55% (626 of 1133) lived within 30 minutes travel time to either significantly longer duration of hospitalization compared with of the hospitals, and 28% (320 of 1133) lived more than 30 suspected cases of enteric fever (Table 2). minutes travel time away. Approximately three quarters of all The majority (89%, 1658 of 1870) of all cases were clinically suspected and laboratory-confirmed enteric fever cases (74%, diagnosed with enteric fever on admission (Table 3). Other 1193 of 1870) identified at the study hospitals were younger frequent diagnoses included gastrointestinal tract infection than 5 years, and 8% (156 of 1870) of enteric fever cases were (n = 57, 3%) and lower respiratory infection (n = 38, 2%). Median infants (<1 year) (Figure 2). Among infants, 90% (141 of hospital duration of the enteric fever cases was 5 (IQR, 4–7) 156) were suspected and 10% (15 of 156) were laboratory-con- days. Hospital duration for confirmed cases (median, 6; IQR, firmed cases. No laboratory-confirmed enteric fever cases were 4–8 days) was longer than that for suspected cases (median,

Table 2. Characteristics of Suspected and Laboratory-Confirmed Enteric Fever Cases, Dhaka, Bangladesh, 2013–2014 (n = 1870)

Suspected Confirmed (n = 1591) (n = 279) Total (n = 1870)

Characteristics n % n % n % P Value Age (median [IQR]), years 3 (1–5) 3 (1–6) 3 (1–5) .05 Gender Female 711 44.7 127 45.5 838 44.8 .79 Male 880 55.3 152 54.5 1032 55.2 Home Address Within Dhaka 1341 84.3 265 95.0 1606 85.9 <.001 Outside Dhaka 250 15.7 14 5.0 264 14.1 Duration of hospitalization (median [IQR]), days 5 (4–7) 6 (4–8) 5 (4–7) <.001 Outcome Discharged 1494 93.9 267 95.7 1761 94.2 Left against medical advice 94 5.9 10 3.6 104 5.6 .003 Referred 0 0.0 2 0.7 2 0.1 Died 3 0.2 0 0.0 3 0.2

Abbreviation: IQR, interquartile range.

S198 • JID 2018:218 (Suppl 4) • Saha et al Table 3. Clinical Diagnoses of Suspected and Laboratory-Confirmed Enteric Fever Cases on Admission, Dhaka, Bangladesh, 2013–2014 (n = 1870)

Suspected Confirmed (n = 1591) (n = 279) Total (n = 1870)

Diagnosis on Admission n % n % n % P Value Enteric fever 1432 90.0 226 81.0 1658 88.7 <.001 Gastrointestinal tract infection 40 3.0 17 6.1 57 3.0 .002 Lower respiratory infection 33 2.0 5 1. 8 38 2.0 .764 Febrile convulsion 13 1. 0 14 5.0 27 1. 4 <.001 Viral fever (including dengue fever and mumps) 15 1. 0 2 0.7 17 0.9 .714 Urinary tract infection 12 1. 0 1 0.4 13 0.7 .463 Central nervous system infection 5 0.0 7 2.5 12 0.6 <.001 Sepsis 9 1. 0 0 0.0 9 0.5 - Hepatitis 3 0.0 2 0.7 5 0.3 .115 Pyrexia of unknown origin 4 0.0 0 0.0 4 0.2 - Upper respiratory infection 2 0.0 0 0.0 2 0.1 - Other 23 1. 4 5 1. 8 28 1. 5 .660

5; IQR, 4–7 days; P < .001). At DSH, enteric fever cases had laboratory-confirmed enteric fever was responsible for approx- longer hospital stay (median, 5; IQR, 4–7 days) compared with imately 4% of all hospitalizations at the 2 largest pediatric hos- that at SSFH (median, 5; IQR, 3–6 days; P < .001). Three deaths pitals in Dhaka, Bangladesh between 2013 and 2014. We also occurred (0.2% of cases) and all were among suspected cases. report that over 60% of culture-positive cases had confirmed Two deceased cases were infants, whereby no blood culture was enteric fever and that Salmonella Typhi was the primary patho- performed, and the third case was a 3-year old child, whose gen identified. Although 3 patients died during hospitalization, blood culture did not yield any microbiological growth. none of these deaths were reported among laboratory-con- Information on the clinical profile of enteric fever cases was firmed typhoid or paratyphoid cases. available for 834 (45%) cases, including 555 (35%) suspected In general, our study finds that a large percentage of infec- and 279 (100%) confirmed cases. All of these cases had history tious diseases in Dhaka can be attributed to enteric fever, of fever (100%), and almost all cases (n = 795, 95%) had high consistent with single-center studies previously conducted in fever (temperature ≥38.9 °C) on the day of admission (Table 4). Bangladesh [14, 17]. For instance, one study collected blood specimens from 103 679 hospitalized and nonhospitalized DISCUSSION patients attending the hospital attached to icddrb, between Enteric fever remains a public health problem in middle- and January 2005 and December 2014, and found that 13.6% of cul- low-income countries. In this study, we report that suspected or tured blood samples were positive, with Salmonella Typhi being

Table 4. Clinical Profile of Suspected and Laboratory-Confirmed Enteric Fever Cases, Dhaka, Bangladesh, 2013–2014 (n = 834)

Suspected (n = 555) Confirmed (n = 279) Total (n = 834)

Variable n % n % n % P Value Symptoms History of fever 555 100.0 279 100.0 834 100.0 - Diarrhea 155 27.9 98 35.1 253 30.3 .033 Vomiting 148 26.7 96 34.4 244 29.3 .02 Abdominal pain 42 7. 6 45 16.1 87 10.4 .0001 Convulsion 31 5.6 20 7. 2 51 6.1 .368 Constipation 23 4.1 8 2.9 31 3.7 .357 Abdominal tenderness 4 0.7 11 3.9 15 1. 8 .002 Headache 5 0.9 0 0.0 5 0.6 - Difficulty in breathing 9 1. 6 0 0.0 9 1. 1 - Blood in stool 2 0.4 1 0.4 3 0.4 .739 Rash 3 0.5 0 0.0 3 0.4 - Maximum Temperature on the Day of Admission (°C) Mean (standard deviation) 39.49 (0.60) 39.52 (0.69) 39.50 (0.63) .238 ≥38.9 399 71.89 212 75.99 611 73.26 38–<38.9 147 26.49 56 20.07 203 24.34 <38 9 1.62 11 3.94 20 2.40

Enteric Fever Among Children of Urban Dhaka • JID 2018:218 (Suppl 4) • S199 the most frequently isolated microorganism (36.9% of positive Supplement sponsorship. This article is part of the supplement blood samples) [14]. “Surveillance for Enteric Fever in Asia Project,” sponsored by the . Among the hospitalized enteric fever cases, the majority were Potential conflicts of interest. S. K. S. has received research grants from younger than 5 years, which is consistent with earlier studies GlaxoSmithKline, Sanofi Pasteur, and Pfizer. All authors have submitted the [6, 11, 13, 18]. The overall isolation rate of Salmonella Typhi ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. and Salmonella Paratyphi A was lower than that reported by 3 studies [11–13]. This may be because the surveillance was not planned to systematically identify enteric fever cases in all pedi- References atric hospitalized patients. The proportion of laboratory-con- 1. Azmatullah A, Qamar FN, Thaver D, Zaidi AK, Bhutta ZA. Systematic review of the global epidemiology, clinical and laboratory profile of enteric fever. J Glob firmed cases was lower in children who lived outside Dhaka Health 2015; 5:020407. compared to that in the children that lived within Dhaka. This 2. Akullian A, Ng’eno E, Matheson AI, et al. Environmental transmission of typhoid fever in an urban slum. PLoS Negl Trop Dis 2015; 9:e0004212. is possibly due to the fact that the cases who come from outside 3. Darton TC, Meiring JE, Tonks S, et al. The STRATAA study protocol: a pro- Dhaka are usually referred from other hospitals and thereby gramme to assess the burden of enteric fever in Bangladesh, Malawi and Nepal using prospective population census, passive surveillance, serological studies and treated with antibiotic before referral. It is likely that this prior healthcare utilisation surveys. BMJ Open 2017; 7:e016283. treatment with antibiotic reduced the chance of yielding a posi- 4. Vos T, Allen C, Arora M, et al. Global, regional, and national incidence, preva- tive growth in blood culture. lence, and years lived with disability for 310 diseases and injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 2016; This study is not without limitations. Reliance on existing 388:1545–602. hospital surveillance databases may lead to missing data on key 5. Wang H, Naghavi M, Allen C, et al. Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980– clinical characteristics. Moreover, the study only collected infor- 2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet mation of cases that were hospitalized at pediatric medicine 2016; 388:1459–544. 6. Saha S, Islam M, Uddin MJ, et al. Integration of enteric fever surveillance into IPDs. Many enteric fever cases receive treatment at outpatient the WHO-coordinated invasive bacterial-vaccine preventable diseases (IB-VPD) departments (OPD) and are generally not hospitalized (Saha S platform: a low cost approach to track an increasingly important disease. PLoS Negl Trop Dis 2017; 11:e0005999. et al, submitted). Some cases might have been admitted to the 7. Luby SP, Saha S, Andrews JR. Towards sustainable public health surveillance for surgery department for complications of enteric fever and were enteric fever. Vaccine 2015; 33(Suppl 3):C3–7. not included in the ongoing surveillance. Inclusion of enteric 8. Kariuki S, Gordon MA, Feasey N, Parry CM. Antimicrobial resistance and man- agement of invasive Salmonella disease. Vaccine 2015; 33(Suppl 3):C21–9. fever cases that were treated at the OPD or other departments 9. Date KA, Newton AE, Medalla F, et al. Changing patterns in enteric fever inci- is likely to provide a more comprehensive overview of the spec- dence and increasing antibiotic resistance of enteric fever isolates in the United States, 2008-2012. Clin Infect Dis 2016; 63:322–9. trum of the disease. The SEAP Phase II is specifically designed 10. Steele AD, Hay Burgess DC, Diaz Z, Carey ME, Zaidi AK. Challenges and oppor- for identification of enteric fever and is expected to address tunities for typhoid fever control: a call for coordinated action. Clin Infect Dis 2016; 62(Suppl 1):S4–8. these limitations by estimating the burden and severity of 11. Naheed A, Ram PK, Brooks WA, et al. Burden of typhoid and paratyphoid fever in enteric fever through comprehensive enteric fever surveillance. a densely populated urban community, Dhaka, Bangladesh. Int J Infect Dis 2010; 14(Suppl 3):e93–9. 12. Brooks WA, Hossain A, Goswami D, et al. Bacteremic typhoid fever in children in CONCLUSIONS an urban slum, Bangladesh. Emerg Infect Dis 2005; 11:326–9. 13. Saha SK, Baqui AH, Hanif M, et al. Typhoid fever in Bangladesh: implications for Despite all of these limitations, our study demonstrated high vaccination policy. Pediatr Infect Dis J 2001; 20:521–4. burden of culture-confirmed enteric fever (Salmonella Typhi or 14. Ahmed D, Nahid MA, Sami AB, et al. Bacterial etiology of bloodstream and antimicrobial resistance in Dhaka, Bangladesh, 2005-2014. Antimicrob Resist Salmonella Paratyphi A) at IPDs of 2 pediatric hospitals (DSH Infect Control 2017; 6:2. and SSFH) in Dhaka, Bangladesh. Enrollment of cases from 15. Crump JA, Ram PK, Gupta SK, Miller MA, Mintz ED. Part I. Analysis of data gaps pertaining to Salmonella enterica serotype Typhi infections in low and diverse departments (OPD and IPD) of these 2 hospitals will medium human development index countries, 1984-2005. Epidemiol Infect 2008; likely enhance the generalizability of enteric fever incidence cal- 136:436–48. 16. Saha SK, Naheed A, Arifeen El S, et al. Surveillance for invasive Streptococcus pneu- culations and make the SEAP Phase II a success. moniae disease among hospitalized children in Bangladesh: antimicrobial suscepti- bility and serotype distribution. Clin Infect Dis. 2009; 48 suppl 2(s2):S75–81. Notes 17. Maude RR, Ghose A, Samad R, et al. A prospective study of the importance of enteric fever as a cause of non-malarial febrile illness in patients admitted Financial support. This work was supported by a grant from the Bill to Chittagong Medical College Hospital, Bangladesh. BMC Infect Dis 2016; and Melinda Gates Foundation (OPP11130007) and the World Health 16:567. Organization (201588766, 201233523, 201022732, 200749550). This article 18. Dewan AM, Corner R, Hashizume M, Ongee ET. Typhoid fever and its association is part of the supplement “Surveillance of Enteric Fever in Asia Project,” with environmental factors in the Dhaka Metropolitan Area of Bangladesh: a spa- sponsored by the Sabin Vaccine Institute. tial and time-series approach. PLoS Negl Trop Dis 2013; 7:e1998.

S200 • JID 2018:218 (Suppl 4) • Saha et al